Autopoiesis, Bioelectric Cognitive Glue, and the Markovian Boundary: A Scale-Free Synthesis of Maturana, Levin, and Active Inference
Year: 2026
Type: article
Abstract: Contemporary theoretical biology faces a fundamental tension between stochastic computational functionalism—which asserts the substrate-independent nature of mind—and biological naturalism, which argues that sentient experience is inherently tied to living matter. In this paper, we resolve this apparent contradiction through an isomorphic synthesis of three foundational theoretical frameworks: the autopoiesis of Humberto Maturana and Francisco Varela, the bioelectric basal cognition and Technological Approach to Mind Everywhere (TAME) of Michael Levin, and Active Inference under the Free Energy Principle formulated by Karl Friston. We demonstrate that biological autopoiesis is not a blind metabolic mechanism, but rather the physical instantiation of a dynamic Markov Blanket whose semipermeable boundary preserves operational closure. At the tissue level, we establish that transmembrane potential gradients (Vmem) and intercellular gap junctions act as the "cognitive glue" that dissolves individual cellular boundaries through leaky stress diffusivity and engrammatic memory anonymization. This topological fusion expands the Cognitive Light Cone (CLC) from unicellular metabolic homeostasis toward goal-directed navigation in morphogenetic anatomical space. Furthermore, we show that when gap-junctional communication collapses and membrane potential depolarizes (Vmem -> 0 mV), the Markov Blanket ruptures, driving oncogenic defection and somatic amnesia (cancer). Finally, we extend this architecture to silicon substrates, demonstrating that emulating biological agency does not require stochastic hallucination, but rather the deployment of a Kernel Level 0 Deterministic Engine with formal Fail-Closed verification (VOID_AB_INITIO), serving as an immutable interface for ingressing agential patterns from latent space.
Author: Prada Forero, Manuel Enrique
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7225453629